Model Representation of Last Decade Regional Changes of Arctic Snow on Sea ice

Abstract:

Present changes that Arctic snow on sea ice experience due to a warming climate have important implications to the sea ice component, precipitation, heat and radiation budgets. In this study, we analyzed the regional distribution and changes, from 2000 to 2013, of Arctic snow depth simulated with a coupled sea ice-general circulation model. For validation, we compared the modeled snow depths (h_{s_mod}) with airborne snow depth measurements from NASA’s Operation IceBridge (h_{s_OIB}) from 2009 to 2013. As in many current sea-ice models, our model configuration consist on a single-layer snow scheme and lack of explicit snow redistribution processes. The snow is accumulated proportionally to the prescribed sea-ice thickness distribution. Despite the simple scheme, our results show that the h_{s_mod} latitudinal distribution in the western Arctic is in good agreement to the OIB observations. The h_{s_mod} is generally thicker than h_{s_OIB}: for latitudes dominated by first-year ice (between 67° N and 76° N) h_{s_mod} is on average 1.1±7.9 cm thicker than h_{s_OIB}, while for multi-year ice dominated latitudes (> 76° N), h_{s_mod} is on average 3.0±8.8 cm thicker than h_{s_OIB}. By 2013, the Arctic-wide h_s decreased 21 % with respect to the h_s multi-annual mean (2000 to 2013) occurring mainly in first-year ice dominated areas. In a simple snow mass budget, our results show that 65 % of the yearly accumulated snow is lost by sublimation and snowmelt due to the heat transfer between the snow/ice interface and the atmosphere. Despite the yearly recovery of snow in winter, the long-term reduction in the summer sea-ice extent ultimately affects the maximum accumulation of snow in spring. Compared to snow reduction estimates from snow radar measurements, the model results underestimate this loss, and we suggest that this is partially due to the lack of explicit snow redistribution processes in the model, ratifying the need to include these in current sea-ice models to improve the snow representations.